ALLOPOLYPLOIDY 



reduced to 1-36 and some have no chiasmata; that is to say 

 metaphase pairing fails but the reduction of pairing is by no means 

 so large as the reduction of chiasma frequency (Table 15). 



TABLE 15 



CHIASMA FREQUENCIES IN THE PARENTAL SPECIES 

 AND IN THE DIPLOID AND TETRAPLOID HYBRID, 

 PRIMULA KEWENSIS, x = 9 (UPCOTT, 1939) 



I, pairs of univalents; II, bivalents; III + I, trivalents and univalents; IV, 

 quadrivalents. 



Structural differences are too small to show in this hybrid except 

 by their interference with the formation of chiasmata. Segregation 

 is therefore nearly regular. Nevertheless sterility is complete. Each 

 gamete contains an apparently normal set of nine chromosomes, but 

 some are from one parent and some from the other. These sets 

 must be unbalanced and fail to work merely because they are mixed. 

 How this might happen we can now see quite easily. If translocations 

 had occurred between the chromosomes since the separation of the 

 parental species then recombinations of their chromosomes would 

 give duplications and deficiencies which would kill the gametes. 



Now, the vegetatively propagated clone of the diploid Primula 

 kewensis has produced tetraploid shoots on several occasions. Failure 

 of mitosis has led to a doubling of the chromosomes in a cell of the 

 growing point. The new tetraploid plant is fertile and true-breeding 



137 



